An approach for quality of service adaptation in service‐oriented Grids

Abstract

SUMMARYSome applications utilizing Grid computing infrastructure require the simultaneous allocation of resources, such as compute servers, networks, memory, disk storage and other specialized resources. Collaborative working and visualization is one example of such applications. In this context, quality of service (QoS) is related to Grid services, and not just to the network connecting these services. With the emerging interest in service-oriented Grids, resources may be advertised and traded as services based on a service level agreement (SLA). Such a SLA must include both general and technical specifications, including pricing policy and properties of the resources required to execute the service, to ensure QoS requirements are satisfied. An approach for QoS adaptation is presented to enable the dynamic adjustment of behavior of an application based on changes in the pre-defined SLA. The approach is particularly useful if workload or network traffic changes in unpredictable ways during an active session.

Wireless Communication Protocols for Distributed Computing Environments

“…In order to make it possible a job must be mapped in a set of nodes with enough computing power; they can be identified only through an efficient resource discovery technique. In (Al-ali et al, 2003;Sundararaj et al, 2004) all these aspects have been analyzed for networks at "Internet scale"; in particular, the first one proposes an extension of the classical Globus model (Morohoshi & Huang, 2005) adopting a DiffServ (Park, 2006) approach and the second one addresses these problems building a virtual overlay network interconnecting all computing resources in order to route better the network traffic and making the grid infrastructure able to react to bandwidth fluctuations. In the literature the QoS problems relative to the pervasive grid paradigm have been addressed in the following ways:…”

QoS management in service-oriented architectures

“…QoS in Grid computing was studied in GARA [6,7], and G-QoSM [1]. In GARA [6,7], resource reservation and actual allocation are separated in order to support advanced resource reservation for critical requests.…”

“…In GARA [6,7], resource reservation and actual allocation are separated in order to support advanced resource reservation for critical requests. G-QoSM [1] extended a UDDI registry. A QoS broker is introduced and is responsible for QoS negotiation and adaptation.…”

GEMLCA: Running Legacy Code Applications as Grid Services

“…The Grid broker has to make selection and scheduling decisions based on QoS parameters such as availability, cost, performance, response time, security etc. The QoS requirements [4], [11] should be defined by Service Level Agreements (SLA) [24] to guarantee the service delivery. The research team is investigating the performance parameters and requirements of Grid QoS, methods to generate, store and retrieve QoS data of Grid services, and approaches to negotiation and management of SLAs.…”